Key facts
- Drug / target: semaglutide 2.4 mg once weekly, subcutaneous — a GLP-1 receptor agonist.
- Sponsor: Novo Nordisk.
- Design: randomized, double-blind, placebo-controlled, 68-week phase 3a withdrawal study (20-week run-in, then randomized maintenance).
- Population: adults with BMI ≥ 30 (or ≥ 27 with ≥ 1 weight-related comorbidity), without diabetes; 902 entered run-in, 803 were randomized.
- Randomization: 2:1 to continued semaglutide (n=535) or switch to placebo (n=268), plus lifestyle intervention in both arms.
- Primary endpoint: percent change in body weight from week 20 to week 68.
- Headline result: −7.9% (continued) vs +6.9% (placebo); difference −14.8 percentage points (95% CI −16.0 to −13.5; P<0.001).
- Registration: NCT03548987 — ClinicalTrials.gov.
- Publication: JAMA, 2021 — PMID 33755728.
What was the STEP 4 trial?
STEP 4 was a randomized, double-blind, placebo-controlled phase 3a trial sponsored by Novo Nordisk and published in JAMA in 2021 (PMID 33755728). It set out to answer a question the earlier weight-loss studies left open: once someone has lost weight on semaglutide, does continuing the drug matter — or can they stop and hold the result?
It was run at 73 sites across 10 countries between June 2018 and March 2020, and it is registered as NCT03548987 on ClinicalTrials.gov. Rather than compare drug against placebo from day one, STEP 4 used a withdrawal design — a structure built specifically to test what continuing versus stopping does to weight already lost.
The withdrawal design
The trial had two stages. In the first, every participant received once-weekly subcutaneous semaglutide during a 20-week run-in: 16 weeks of gradual dose escalation followed by 4 weeks at the 2.4-mg maintenance dose. This was a shared starting point — nobody was on placebo yet.
Only participants who tolerated treatment and reached the 2.4-mg maintenance dose moved into the second stage. There they were randomly assigned, in a 2:1 ratio, either to continue semaglutide or to switch to a matching placebo, with lifestyle intervention in both arms, for a further 48 weeks (week 20 to week 68). Because the study was double-blind, neither participants nor investigators knew who had been switched.
The value of this run-in-then-randomize approach is that it compares two paths for people who are all in the same place — established on the drug and having lost weight. The only variable that changes at randomization is whether the medication continues.
Who was in it?
STEP 4 enrolled adults with a body-mass index of at least 30, or at least 27 with one or more weight-related comorbidities, and without diabetes. A total of 902 participants started the semaglutide run-in. After 20 weeks, 803 of them (89.0%) had reached the 2.4-mg maintenance dose and were randomized — 535 to continued semaglutide and 268 to placebo.
At randomization the group had a mean age of 46 years, was 79% women (634 of 803), and had a mean body weight of 107.2 kg. During the run-in, participants had already lost a mean of 10.6% of their body weight. Retention was high: 787 participants (98.0%) completed the trial and 741 (92.3%) completed treatment.
What happened after randomization
The primary endpoint was the percent change in body weight from week 20 (the point of randomization) to week 68. The two groups moved in opposite directions.
Participants who continued semaglutide kept losing weight, with a mean change of −7.9% over that window. Participants switched to placebo regained weight, with a mean change of +6.9%. The difference between the arms was 14.8 percentage points (95% CI −16.0 to −13.5; P<0.001).
| Body-weight change, week 20 → week 68 | Continued semaglutide (n=535) | Switched to placebo (n=268) | Difference |
|---|---|---|---|
| Mean percent change in body weight (primary endpoint) | −7.9% | +6.9% | −14.8 pp (95% CI −16.0 to −13.5); P<0.001 |
Read plainly: the people who stopped the drug did not simply stall at the weight they had reached — they trended back upward, while the people who stayed on it continued downward. The headline gap is the sum of those two opposite movements.
Secondary outcomes
The confirmatory secondary endpoints followed the same pattern, all favoring continued semaglutide over placebo. Waist circumference changed by −9.7 cm (95% CI −10.9 to −8.5), systolic blood pressure by −3.9 mm Hg (95% CI −5.8 to −2.0), and the SF-36 physical-functioning score by 2.5 points (95% CI 1.6 to 3.3), each with P<0.001. In other words, the cardiometabolic and functional measures moved in step with the weight difference. For how much weight semaglutide produces from a placebo-controlled start, see our summary of the STEP 1 trial, and for the broader class picture, peptides for weight loss.
Safety reported
The tolerability signal was consistent with the known profile of GLP-1 receptor agonists. Gastrointestinal events were reported in 49.1% of participants who continued semaglutide versus 26.1% of those on placebo. Despite that gap, treatment discontinuation because of adverse events was similar in the two groups — 2.4% with continued semaglutide and 2.2% with placebo. This page reports only the figures stated in the published abstract rather than attaching rates to individual symptoms.
One point to keep in view when reading these numbers: everyone had already been through the escalation phase during the run-in, which is when gastrointestinal effects in this drug class are typically most pronounced. The randomized phase therefore reflects tolerability in people already established on the maintenance dose.
What STEP 4 does and does not prove
STEP 4 gives a clean answer to a narrow question, and it helps to keep both halves in mind:
- What it shows: among people who lost weight on semaglutide and reached the maintenance dose, continuing the drug preserved and extended that loss, whereas stopping led to regain. That is direct evidence about continuation versus withdrawal.
- It studied responders. Only participants who reached the 2.4-mg dose during run-in were randomized. The trial describes what happens to that established group, not to everyone who ever starts the drug.
- It is 68 weeks, not a lifetime. The trial followed people to week 68. It does not describe outcomes over many years, nor what a slower taper or a different maintenance strategy might do.
- It is one drug versus its own placebo. STEP 4 does not rank semaglutide against other agents, and it does not isolate a mechanism for the regain — it reports a clinical outcome.
None of this weakens the core finding. It keeps it the right size: strong, specific evidence that this treatment's benefit is tied to staying on it.
Why STEP 4 matters
STEP 4 is the trial most often cited for the idea that obesity behaves like a chronic condition rather than something a fixed course of medication cures. When the drug was withdrawn, the body did not hold the new set point — weight moved back up. That reframes the practical question from "how much can I lose?" to "am I prepared to treat this on an ongoing basis?", a conversation that belongs with a clinician.
For anyone actually on a weekly injection, the study also underlines why consistency and record-keeping matter: the difference between the two arms was simply whether treatment continued. Keeping an accurate log of doses, dates, and how you feel makes those appointments run on data — which is exactly what tracking a peptide or medication properly is for. For the wider evidence map of which weight-loss drugs carry trials of this caliber, see our pillar guide on peptides for weight loss, and the companion cardiovascular-outcomes summary in the SELECT trial.
Frequently asked questions
What did the STEP 4 trial show?
STEP 4 showed that continuing semaglutide 2.4 mg maintained and extended weight loss, while switching to placebo led to weight regain. After a 20-week run-in in which participants lost a mean of 10.6% of body weight, those who continued semaglutide lost a further 7.9% from week 20 to week 68, whereas those switched to placebo regained 6.9% — a difference of 14.8 percentage points (95% CI -16.0 to -13.5; P less than 0.001).
What is a withdrawal or run-in trial design?
In STEP 4 everyone first received semaglutide for a 20-week run-in period (16 weeks of dose escalation and 4 weeks at the 2.4-mg maintenance dose). Only participants who responded and reached the target dose then entered the randomized phase, where they were assigned 2:1 to keep taking semaglutide or switch to placebo. This design isolates the effect of continuing versus stopping the drug in people already established on treatment.
How much weight did people regain after stopping semaglutide?
Participants switched to placebo regained weight, with a mean body-weight change of +6.9% from week 20 to week 68. Those who continued semaglutide instead lost a further 7.9% over the same period. The gap between the two groups was 14.8 percentage points (95% CI -16.0 to -13.5; P less than 0.001).
Does STEP 4 mean obesity needs ongoing treatment?
STEP 4 supports the view that, for these medications, weight loss depends on continued treatment: when the drug was withdrawn, weight moved back up. That is consistent with obesity being treated as a chronic condition rather than cured by a fixed course. STEP 4 tested one drug in one population over 68 weeks, so it describes what happened in that trial rather than dictating any individual's care, which remains a clinical decision.
What dose of semaglutide was used in STEP 4?
During the run-in, semaglutide was titrated over 16 weeks to a once-weekly subcutaneous maintenance dose of 2.4 mg, followed by 4 weeks at that dose. In the randomized phase participants either continued semaglutide 2.4 mg weekly or received a matching placebo. This reports what the trial administered; it is not a dosing recommendation, and semaglutide is a prescription decision made with a clinician.
Is STEP 4 the same as STEP 1?
No. STEP 1 measured how much weight semaglutide produces against placebo from the start of treatment. STEP 4 is a withdrawal study: everyone started on semaglutide, and only after a 20-week run-in were responders randomized to continue or stop. STEP 1 answers how well semaglutide works; STEP 4 answers what happens when you keep taking it versus stop.
Sources
This summary traces to the peer-reviewed publication and trial registration:
- Effect of Continued Weekly Subcutaneous Semaglutide vs Placebo on Weight Loss Maintenance in Adults With Overweight or Obesity: The STEP 4 Randomized Clinical Trial — JAMA, 2021;325(14):1414–1425; PubMed 33755728.
- STEP 4 (semaglutide 2.4 mg weight maintenance) — ClinicalTrials.gov NCT03548987.
- Drug reference record: semaglutide.